Retroviruses, including HIV and HTLV, cause neurological disease. The goal of this project is to understand how retroviruses adversely affect the central nervous system. We are utilizing mouse retroviruses which cause rapid neurologic diseases (ie. 2-3 weeks after inoculation) and have used this system to study the interaction between virus and brain. Entry of virus into the brain: A non-structural glycosylated form of the viral gag polyprotein (glycogag) has long been recognized to be highly conserved among murine, feline and some primate retroviruses, but its function is as yet unknown. We have found that expression of this protein is a critical determinant neuroinvasiveness, and that this protein functions to enhance the spread of virus in vivo. Nonpathogenic mutant viruses which lack this protein rapidly generate revertants which have regained their pathogenicity for the brain. Sequence analysis of the revertants reveals a consistent mutation which generates a new start site for translation producing a slightly truncated protein. This indicates strong selective pressure favoring viruses which express glycogag. We are interested in understanding its function at the molecular level. In vitro model of neurodegeneration: We are continuing our efforts to develop a system whereby one can study at the molecular level the mechanism of retrovirus-induced neurodegeneration. We have initiated a study of embryonic cortical neurons cultured on a glial bed. This system has been used successfully by others to study neurotoxicity induced by excitatory neurotransmitters as well as hypoxia. These neurons are relevant since they are also susceptible to degeneration in virus- infected mice. Preliminary results suggest that supernatants from retrovirus-infected but not from uninfected pure microglial cultures cause neurotoxicity in this system. We plan to extend these studies to the analysis of microglia expressing individual viral genes and have constructed the appropriate expression vectors. Envelope gene: Two chimeric murine retroviruses which utilize the same receptor but differ in their envelope sequences have both been shown to infect the brain, and appear also to infect the same cell types in the brain. Both induce an intense diffuse astrocytosis, but only one causes clinical neurologic signs. We are very interested in this paradigm because of its analogy to brain infection by HIV, in which HIV envelope sequences of macrophage tropic viruses have been recovered from both demented and non-demented AIDS patients. We are currently looking for morphologic correlates of neurovirulence in these mice and are currently analyzing cytokine profiles in the brain.